Machine Learning Approach for Ex-Post Evaluation of Road Traffic Collision Severity Trends
Bibliographic record
Abstract
Road traffic collisions (RTCs) have been a key concern because of their negative impact on road users’ safety and social aspects. To address this issue, many research efforts have been made, yet there is still very little known about the various key factors affecting RTC severity and the ability to feasibly provide the overall RTC severity classification trends. This study aimed to quantify the factors affecting the severity of RTCs using neural network search and sensitivity analysis. To this end, an ex-post evaluation framework that could systematically classify RTC severity trends in two stages was developed: stage 1 involved RTC severity classification based on a radial basis-function-driven machine learning model, and stage 2 utilized global sensitivity analysis to identify critical factors affecting RTC severity trends. It was shown that the radial basis function network models accurately predicted the RTC severity (with 77% accuracy) based on multi-contextual inputs and derived three key factors (road classification, number of vehicles involved, and speed limit) associated with the severity based on more than 12,000 RTC records collected over 6 years in the county of Cambridgeshire, UK. Generalized RTC severity trends associated with the results were also proposed and discussed as an ex-post evaluation. The outcomes of this study will help transportation authorities and engineers by serving as a benchmark and predictable reference to minimize the negative impact of potential RTCs on road users’ safety as well as social costs.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.006 | 0.013 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".